Leap Seconds

stevebrassett

Well-known member
Joined
26 Jul 2004
Messages
3,574
Location
Herts
Visit site
If the leap second is the solution to the problem and we remove the solution without addressing the problem, aren't we causing ourselves trouble?
 

mjcoon

Well-known member
Joined
18 Jun 2011
Messages
4,635
Location
Berkshire, UK
www.mjcoon.plus.com
If the leap second is the solution to the problem and we remove the solution without addressing the problem, aren't we causing ourselves trouble?

Did you read (meaning, understand!) the pros and cons? It seems to me that if we have had a couple of dozen leap seconds already, those complaining they are hard work have just failed to produce a proper procedure for dealing with them.

When I got my first GPS back in the 1990s I amused myself by watching it cope with leap seconds at the correct midnights. The dates were announced in advance so I could do this. The GPS signals tell devices when leap seconds occur. No doubt this is one of the reasons why GPS is used as a source of timing for non-navigational purposes.

Mike.
 

stevebrassett

Well-known member
Joined
26 Jul 2004
Messages
3,574
Location
Herts
Visit site
Yes I did read and understand the pros and cons. They simply listed the pros and cons of having leap seconds and didn't really suggest a suitable alternative. Changing timezone is not a suitable alternative.
 

DJE

Well-known member
Joined
21 Jun 2004
Messages
7,666
Location
Fareham
www.casl.uk.com
Yes I did read and understand the pros and cons. They simply listed the pros and cons of having leap seconds and didn't really suggest a suitable alternative. Changing timezone is not a suitable alternative.

I don't suppose there are many people using astro navigation for real these days but for those who are couldn't we seperate UTC and GMT? So that UTC kept time with the atomic clocks and GMT was adjusted to keep pace with the earth's rotation? The Royal Observatory would have to publish a correction once a year advising the difference between the two.
 

laika

Well-known member
Joined
6 Apr 2011
Messages
8,206
Location
London / Gosport
Visit site
I don't suppose there are many people using astro navigation for real these days but for those who are couldn't we seperate UTC and GMT? So that UTC kept time with the atomic clocks and GMT was adjusted to keep pace with the earth's rotation? The Royal Observatory would have to publish a correction once a year advising the difference between the two.

Pretty much, yes. If you check your almanac you'll see the times are actually "UT" which is based on the earth's rotation like GMT. UTC is based on SI seconds, ticking with atomic clocks but with leap seconds added in to keep it within 0.9 seconds of UT. So UTC is distinct from UT but never by enough to make much of a difference. The wikipedia page has more info than most people care about on the finer points of different types of UT. I would guess the same thing as you: if they abandon leap seconds: they'd need to publish a correction in the almanac and we'd have an additional step, albeit a trivial one, in our calculations.
 

lpdsn

New member
Joined
3 Apr 2009
Messages
5,467
Visit site
When I got my first GPS back in the 1990s I amused myself by watching it cope with leap seconds at the correct midnights. The dates were announced in advance so I could do this. The GPS signals tell devices when leap seconds occur. No doubt this is one of the reasons why GPS is used as a source of timing for non-navigational purposes.

Mike.

Interesting. I reviewed a document at work about six months ago regarding using GPS for our NTP servers. The GPS time (can't remember the name off the top of my head) doesn't support leap seconds so has drifted away from UTC and the NTP servers had to handle the offset. I guess you're saying that there is a signal that can be sent to change the value of the offset.
 

laika

Well-known member
Joined
6 Apr 2011
Messages
8,206
Location
London / Gosport
Visit site
The GPS time (can't remember the name off the top of my head) doesn't support leap seconds so has drifted away from UTC and the NTP servers had to handle the offset. I guess you're saying that there is a signal that can be sent to change the value of the offset.

Offset from UTC is sent as part of the GPS navigational message:
http://en.wikipedia.org/wiki/GPS_signals#Navigation_message

I think whenever this topic comes up there are people who hear that GPS time is different from UTC and worry that the time shown on their plotters is not the "real" time. In fact what is displayed is invariably UTC: The offset will have been applied (as it would be with any NTP servers). If you check out the structure of GPS-related NMEA-0183 sentences such as RMC they contain UTC rather than GPS time.
 

Buck Turgidson

Well-known member
Joined
10 Apr 2012
Messages
3,441
Location
Zürich
Visit site
Is it me or is the RYA report factually incorrect. It omits any mention of TIA which is the basis of SI time measurements. TIA is ahead of UTC by 35 seconds as these are the leap seconds inserted to UTC since 1958. GPS time was synchronised to UTC in 1980 and is therefore always 19 seconds behind TIA. UTC is a further 16 seconds behind. And as has been said, the offset to UTC is transmitted in the GPS almanac message and should be corrected by the receiver.
 
Top